[llvm] [AMDGPU] Account for inline asm size in inst_pref_size calculation (PR #192306)

[Adel Ejjeh via llvm-commits]

================
@@ -0,0 +1,62 @@
+//===- llvm/unittests/Target/AMDGPU/InstCacheLineSize.cpp -----------------===//
----------------
adelejjeh wrote:

The place we invoke getInstCacheLineSize is in evalueateInstPrefSize which we never currently exercise for pre-gfx11 architectures, so we never test the path that takes the if(FeatureInstCacheLineSize64). No other pass invokes this so I don't see how to lit-test it.

https://github.com/llvm/llvm-project/pull/192306